https://github.com/jeongukjae/tf-serving-s3
Custom TensorFlow Serving image with AWS S3 filesystem support
https://github.com/jeongukjae/tf-serving-s3
Last synced: 4 days ago
JSON representation
Custom TensorFlow Serving image with AWS S3 filesystem support
- Host: GitHub
- URL: https://github.com/jeongukjae/tf-serving-s3
- Owner: jeongukjae
- Created: 2023-01-17T14:44:52.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-01-17T17:30:55.000Z (over 2 years ago)
- Last Synced: 2025-04-10T14:26:25.100Z (6 months ago)
- Language: Dockerfile
- Size: 6.84 KB
- Stars: 9
- Watchers: 1
- Forks: 6
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# tf-serving-s3
Custom TensorFlow Serving image built with TensorFlow IO for AWS S3 filesystem support.
**Check [GitHub Packages](https://github.com/jeongukjae/tf-serving-s3/pkgs/container/tf-serving-s3) for the Docker images.**
## Usage
The usage is exactly the same as the official TensorFlow Serving image.
For more details, please refer to the [TensorFlow Serving with Docker](https://www.tensorflow.org/tfx/serving/docker).
But you need to use the image from GitHub Packages instead of the official image.
```bash
$ docker run \
-p 8500:8500 \
-p 8501:8501 \
-e AWS_ACCESS_KEY_ID \
-e AWS_SECRET_ACCESS_KEY \
-e AWS_DEFAULT_REGION \
-e MODEL_BASE_PATH=s3://BUCKET_NAME/PATH_TO_MODEL \
-e MODEL_NAME=MODEL_NAME \
ghcr.io/jeongukjae/tf-serving-s3:2.11.0
```
## What is the differences/details?
Check this blog post:
## How to build the image?
```bash
docker build -t ghcr.io/jeongukjae/tf-serving-s3:2.11.0-devel -f Dockerfile.devel .
docker build -t ghcr.io/jeongukjae/tf-serving-s3:2.11.0 -f Dockerfile .
```